Influence Attitude and Behavioral Intention of the Millenial Generation to Adoption of Telemedicine Platforms in Bali in the New Normal Era

.


A B S T R A C T
The pandemic has brought various impacts and changes in people's lives, one of which is the use of digital telemedicine applications. Health is a vital aspect of human life; therefore, telemedicine applications have an important role in facing the changing era. This study attempts to describe the attitudes and behavioral intentions of the millennial generation in adopting telemedicine platforms in the new normal era. The data in this study were collected by distributing questionnaires to 120 respondents, namely the millennial generation in Bali. Data were analyzed using the SEM-PLS analysis technique. The results of this study found that perceived ease of use, social influence, perceived usefulness had a positive and significant effect on behavioral intention, while e-trust had no effect. Perceived ease of use has a positive and significant effect on perceived usefulness. Social influence has a positive and significant effect on E-trust. E-trust is not able to mediate the influence of social influence on behavioral intention. Perceived usefulness is able to positively mediate the effect of perceived ease of use on behavioral intention. This research is expected to provide an overview of consumer behavior towards digital services, as well as an evaluation for digital service providers themselves.

INTRODUCTION
The emergence of the Covid-19 pandemic and the enactment of this new normal policy have caused various changes to occur, one of which is the acceleration of technological development in Indonesia. Technological developments have provided various kinds of changes in human life, one of which is the development of smartphones and the applications that accompany them (Bhowmick & Hazarika, 2017;Negash et al., 2018). The use of smartphones in the digital era has become very rapid along with the emergence of various kinds of applications that help humans in carrying out their activities. These activities IJSSB. P-ISSN: 2614-6533 E-ISSN: 2549-6409 include business and marketing, health, recreation, education, finance and so on. This development is supported by high internet users in Indonesia, thus providing opportunities for business players to develop. As reported by the Indonesian Internet Providers Association (2019), Indonesia has 107.2 million Internet users, an increase of 12.6% compared to 2018. In the second quarter of 2020, it increased to 73.7 percent of Indonesia's population, equivalent to 196.7 million users (Nisa & Solekah, 2022).
The number of internet users in Indonesia has experienced quite significant growth. This condition certainly provides high opportunities for business actors in the digital world to continue to develop and meet the needs of society. The health industry, which is one of the busiest industries during a pandemic and is at the forefront of fighting the Covid-19 virus, should also maximize digital technology so that it can help provide health services. This is due to the influence of new habits from the new normal policy which has resulted in many thinking people not coming directly to the hospital for treatment because they are considered to have the potential to become one of the clusters for the spread of the virus due to the high mobility of patients in the hospital and the treatment of suspected patients, probable, or confirmed COVID-19. In addition, some people also don't want to come to the hospital because they are afraid of being diagnosed with Covid. To answer this problem, one of the digital technology innovations that can help health services in the new normal era is the telemedicine platform.
Telemedicine is defined as the use in the field of communication and information technology which is carried out electronically with the aim of providing and supporting patient health care without any separation (Fong et al., 2020;Gunter et al., 2016). Telemedicine is the practice of health using audio, visual and data communication, including care, diagnosis, consultation and treatment as well as the exchange of medical data and scientific discussions remotely. Based on the above understanding, it can be understood that the scope of telemedicine is quite broad, covering the provision of remote health services (including clinical, educational and administrative services) through the transfer of information (audio, video, graphics), using telecommunications devices (interactive audio-video). two-way, computer and telemetry) involving doctors, patients and other parties (Sari & Wirman, 2021). There are two main concepts in telemedicine, namely the real time method and store and forward (Chandwani & Dwivedi, 2015).
In many countries, telemedicine is a strategy to prevent the spread of COVID-19 due to its nature of providing health services using electronic communication technology so that patients and medical personnel do not need to meet in person in one place but can communicate through an application (Chang & Boudier-Revéret, 2020;Song et al., 2020). Referring to a survey from the Association of Indonesian Internet Service Providers (APJII) in 2017, 51.06% provide benefits of the internet in the health sector, provide benefits in digging up information about health, while 14.05% provide benefits for consulting with health experts. (APJII, 2017). These data show that the use of the internet in the health sector has occurred even before the Covid-19 pandemic, so it can be indicated that the use of telemidicine in the new normal era will also increase.
There are various digital health applications (telemedicine) currently developing in Indonesia, the results of a survey conducted by katadata.com in 2021 to find out the use of e-health applications by the public during the new normal era. Halodoc takes the lead with an assessment of applications that have been used. Telemedicine already has its own market share, especially in the face of a pandemic. One of the most frequently used is Halodoc. In terms of Halodoc, the total visits of health service users increased by 600% during the COVID-19 period (Hermawan & Paramita, 2021;Khotimah et al., 2022). The majority of users who access this platform are the millennial generation who are aged 20-30 years. The data in Figure 2 shows that the Halodoc application is a telemedicine application that has succeeded in becoming top of mind for current telemedicine application users in Indonesia. The high number of users of the Halodoc application can provide a comprehensive picture related to behavioral intention in using the telemedicine application itself, so that Halodoc is used as an object in this study.
The government has intensified the use of telemedicine by issuing a circular from the Minister of Health No. 303/2020, regarding the provision of health services by utilizing communication and information technology through telemedicine in order to prevent the spread of the corona (Drake et al., 2022;Rizzi et al., 2020). However, in reality the implementation of telemedicine is not easy in Indonesia, the resulting failure rate is 70%, which is the fundamental factor, namely not due to failures in the field of technology but due to the lack of readiness for the use of existing technology.
The use of telemedicine technology such as Halodoc still faces challenges for users who are in the millennial generation category, even though it has become top of mind, Halodoc still has to be able to attract other users to use their health services. This is because the millennial generation are people who are more critical and tend to show skepticism about anything new (Widiasih & Darma, 2021). The millennial generation tends to find it easier to compare one aspect with another, so that the use of telemedicine and conventional medicine is still something that needs to be studied more deeply by the millennial generation. Based on the results of an initial survey of 20 millennials in Bali, behavioral intention in using the Halodoc application is still experiencing problems, reflected in 55 percent of respondents stating that their attitudes are still doubtful about service, quality, and doctor's diagnoses if they only use digital services without meeting face to face. direct face. Communities on the island of Bali who are still very strong with their culture and local wisdom also create obstacles to using the telemedicine platform because as a system that is technological and long-distance, the application of telemedicine affects the emotional connection and therapeutic touch between doctors and patients because there is no direct interaction. The lack of bond between doctors and patients will sometimes lead to perceptions or distrust of the services provided. In addition, 65 percent of respondents doubted the ease of using Halodoc.
The survey shows that there are several reasons not to use telemedicine applications. There are several reasons someone does not want to use digital health services at this time, including the lack of promotions being carried out, applications that are rarely used, applications that are considered expensive, less popular, do not have complete features, are less innovative., less than optimal service, and so forth. Of course, these constraints can cause behavioral intention to use not to form in application users. Behavioral intention in this case is the intention to use, which is the level of individual intention to carry out certain behaviors or actions (Sweeney et al., 2015). Behavioral intention is one of the behaviors shown by consumers in forming an interest in using a product or a service (Kim & Gambino, 2016;Weng et al., 2018). Behavioral intention in the development of telemedicine needs to be improved, bearing in mind that in the new normal era vigilance against COVID-19 should not be simply removed, so that the use of digital health applications is still recommended in the new normal era. There are several factors that influence behavioral intention, including perceived ease of use and social influence (Prastiawan et al., 2021;Sugihartono et al., 2020;Zhang et al., 2020).
Perceived ease of use is also a factor that is indicated to influence behavioral intention. Previous research stated that perceived convenience is the extent to which an individual believes that using a technology will be free from his efforts (Hermawan & Paramita, 2021). Perceived ease of use usually refers to the user's belief that a technological system used does not require enormous effort when used Syahril & Rikumahu, 2019). Ease of use is very important for Halodoc users, because a situation that requires medical assistance is generally a critical situation, so the ease of use of the application can speed up the process of assistance being provided. Previous research illustrates that perceived convenience has a positive influence on behavioral intention (Beldad & Hegner, 2018;Huei et al., 2018;Weng et al., 2018). Meanwhile, the results of previous research show that perceived ease of use does not affect behavioral intention (Cigdem & Ozturk, 2016).
The social environment in society is also able to encourage an individual to wear something, in this case it is called social influence. Social influence is the influence an individual receives from their social environment, and causes them to do or use something (Afrizal & Wallang, 2021). The influence of a strong social environment can have a positive or negative influence, this influence can lead to interest or vice versa towards a product, especially in the growing digital era, social influences are often encountered in daily activities (Fryt et al., 2021;Lee & Wu, 2018). Results of previous stated that provides an illustration that social influence has a positive influence on behavioral intention (Hu et al., 2019;Jatimoyo et al., 2021). While the results of similar research show that social influence has no effect on behavioral intention (Bimantara, 2021;Nurhaliza, 2022;Wintang & Pasharibu, 2021).
In the use of a technology, the usefulness and convenience alone are felt to be insufficient to increase the behavioral intention to use it. Therefore, other factors are needed as reinforcement in this relationship, namely perceived usefulness and e-trust. Perceived usefulness is able to define as the extent to which a person feels that by using information technology, they feel that there are advantages to be gained, besides that productivity and performance using information technology will increase because work can be completed in not much time (Budiantara et al., 2019;Sarkar et al., 2020). The usefulness of an application is important to increase the behavioral intention of the user. Likewise with health applications such as Halodoc, of course the benefits of using these applications must be felt to the fullest so that behavioral intention to use can be increased from users. Results of previous research illustrates that perceived usefulness can mediate behavioral intention to use a technology (Muliadi & Japarianto, 2021;Subagio & Rachmawati, 2020).
Meanwhile, the trust of a consumer in an e-commerce business is a key factor in carrying out online sales and purchasing activities (Kayali et al., 2019;Tasin, 2017). Here belief refers to the belief of a consumer that the seller is capable of fulfilling his wishes and this belief encourages consumers to complete and accept transactions online (Namahoot & Laohavichien, 2018;Wijaya & Jasfar, 2014). The results of previous research show that e-trust can be a mediating variable in the relationship between aspects of consumer behavior and behavioral intention to use (Jeon et al., 2017;Namahoot & Laohavichien, 2018).
Based on the pandemic and new normal phenomena which have led to the importance of the role of digital health applications in society, it is indicated to be the cause of the use of telemedicine, especially Halodoc, which has shown an increase. However, on the other hand the observations show that the trust of the millennial generation in the Halodoc application is still not optimal, as a result there are still many people who doubt the use of the application. This study attempts to address these issues by analyzing the factors that influence behavioral intention to use telemedicine users.

METHODS
This research was conducted in the Province of Bali with users of telemedicine applications, especially Halodoc as research subjects. Bali Province was used as the research location because Bali is one of the areas heavily affected by the pandemic, and based on observations it was found that there were obstacles in implementing telemedicine applications for the millennial generation in Bali. The population of this study is all millennials in the Province of Bali who use the Halodoc application, which continues to change over time, so the number is not known with certainty. The sample method used is purposive sampling, which is a sample determination method based on certain criteria or considerations (Sugiyono, 2018). The sample was selected based on the following user criteria, (1) Domiciled in Bali Province, (2) Millennials born 1980 -2000 or aged 20 -41 years, (3) Interested in using the Halodoc application to solve problems experienced health. The data is processed using the Structural Equation Modeling (SEM) method according to the SEM rule of thumb to determine the sample is five times to ten times the number of parameters estimated, with the calculation of the number of indicators x 5 = 24 x 5 = 120 respondents.
The variables used in this study are behavioral intention which uses a combination of indicators namely the willingness to use Halodoc when needed, the desire to use Halodoc because it hopes for good performance, the desire to use Halodoc because of the ease of contacting the system, the desire to use Halodoc because of encouragement from the surrounding environment, and the desire to use Halodoc because it has familiar with online and digital systems. The perceived ease of use variable uses a combination of indicators as follows, Halodoc is easy to use/operate, Halodoc is easy to study its features, Features -Halodoc feature has an easy-to-see display, Halodoc makes consultations with doctors easier, Halodoc makes receiving health information easier.
The social influence variable in this study uses indicators as follows, there is influence to use Halodoc from the neighborhood, there is influence to use Halodoc from the family, Using Halodoc because they believe it can increase self-confidence in society, and using Halodoc because there are habits from the surrounding environment to use technology. The perceived usefulness variables in this study use indicators as follows, Using Halodoc increases the effectiveness of health control, Using Halodoc increases user productivity in the health sector, Using Halodoc is useful as a a digital health consultation tool, Using Halodoc promotes a healthy lifestyle, and Using Halodoc reduces anxiety when facing health problems. The e-trust indicator variable in this study uses indicators as follows, believe that Halodoc has credibility in its field, believe that Halodoc provide accurate information, believe that Halodoc simplifies the service process, believe that Halodoc maintains data confidentiality, and believe that Halodoc is able to maintain the integrity of doctors.
This research is classified as a quantitative research type originating from primary data, namely data from the results of distributing questionnaires obtained by distributing questionnaires online to respondents, namely the Balinese millennial generation who have used the Halodoc application. The data collection technique that will be carried out in this study is distributing questionnaires. Questionnaires were distributed using the Google form for millennials in the Province of Bali who had used Halodoc. Questionnaires were distributed to millennials in the Province of Bali who had used Halodoc with a Likert scale of 1-10. Data were analyzed using the Partial Least Squares (PLS) technique. This technique was chosen because the model estimates produced by SEM-PLS generally show a higher level of statistical power and provide similar results in statistical significance and path coefficient estimation (de Sousa-Filho et al., 2020).

Results
Based on the results of the distribution of questionnaires, there were 41 male respondents with a percentage of 34.2%. There were 79 female respondents with a percentage of 65.8%. This shows that female respondents are the most dominating, this is because women have a higher sense of worry about health than men. Respondents aged <20 years were 10 people with a percentage of 8.3%. Respondents aged 20-25 years were 18 people with a percentage of 15%. Respondents aged 26-30 years were 48 people with a I Putu Kresna Saniscara Dewanta Putra / Influence Attitude and Behavioral Intention of the Millenial Generation to Adoption of Telemedicine Platforms in Bali in the New Normal Era percentage of 40%. Respondents aged 31-35 years were 26 people with a percentage of 21.7%. Respondents aged 36-40 years were 13 people with a percentage of 10.8% and respondents aged >40 years were 5 people with a percentage of 4.2%. This shows that respondents aged 26-30 years are the most dominating. All respondents live in Bali, namely as many as 120 people with a percentage of 100%. All respondents in this study were interested in using telemedicine.
In the measurement of the outer model presented in Figure 1, convergent validity, discriminant validity and uni-dimensionality tests were carried out. Convergent validity consists of outer loading and Average Variance Extracted (AVE). Discriminant Validity consists of comparing the outer loading value with the cross-loading value and the AVE root is greater than the correlation between variables. For the reliability test, composite reliability, rho-A and Alpha Cronbach are used (Utama, 2018: 237). The validity of the statistical data used in this study was assessed using convergent and discriminant validity (Adelekan et al., 2018).

Figure 1. Outer Model
In measuring the inner model presented in Figure 2, a direct effect test and an indirect effect test were carried out as well as testing the magnitude of the influence with the analysis of the coefficient of determination (R-Square), analyzing F-Square and Q-square. The structural model or inner model is evaluated by looking at the percentage of variance described, namely by looking at R2 (R-Square exogenous variable) for the dependent latent construct using the Stone-Geisser Q Square test and also looking at the size of the structural path coefficient. Potential mediation will be confirmed after further mediation analysis using the bootstrap method.  Based on Table 1, the R-square value for the variable perceived ease of use and social influence on behavioral intention is 0.717, including the large which shows that it has a large influence of 0.717 x 100% = 71.7%. The R-square value for the variables perceived ease of use and social influence on E-trust is 0.191, which is small which indicates that it has a large influence of 0.191 x 100% = 19.1%. The R-square value for the variables perceived ease of use and social influence on perceived usefulness is 0.280 which is small which indicates that it has a large influence of 0.280 x 100% = 28.0%. Calculation of the Q-square value of 0.835 is more than 0 and close to 1, so that it can be concluded that the model has a predictive relevance value or the model deserves to be said to have relevant predictive value. The results of the Direct Effect Test analysis presented in Table 2 show that the e-trust variable does not have a significant direct influence on behavioral intention, with a coefficient value of -0.037 and a p-value of 0.407. Meanwhile, the perceived ease of use variable has a significant direct influence on behavioral intention, with a coefficient value of 0.572 and a p-value of 0.002. The perceived ease of use variable also has a significant direct influence on perceived usefulness, with a coefficient value of 0.530 and a p-value of 0.003. In addition, the perceived usefulness variable also has a significant direct influence on behavioral intention, with a coefficient value of 0.214 and a p-value of 0.001. Social influence variables have a significant direct influence on behavioral intention, with a coefficient value of 0.216 and a p-value of 0.039. Finally, social influence variables also have a significant direct influence on e-trust, with a coefficient value of 0.437 and a p-value of 0.000. The results of the analysis presented in Table 3 state that social influence has a significant indirect influence on behavioral intention through e-trust mediators, where the indirect influence coefficient is -0.016. However, the t-statistics value was only 1,100 and the p-value was 0.386, which showed no significant effect of social influence on behavioral intention through e-trust mediators. Meanwhile, in the second analysis, perceived ease of use has a significant indirect influence on behavioral intention through mediator perceived usefulness, where the indirect influence coefficient is 0.113. Very large t-statistics values (56,538) and very small p-values (0.000) indicate that the indirect effect of perceived ease of use on behavioral intention through mediator perceived usefulness is very significant. From the results of the analysis, it can be concluded that perceived ease of use has a greater influence on behavioral intention than social influence through their respective mediators (perceived usefulness and e-trust).

Discussion
Based on the results of data analysis, it was obtained that the p-value of the variable perceived ease of use for behavioral intention was 0.002 compared to a significant value of 0.05. Because the p-value < significant (0.002 < 0.05) with a beta value of 0.572 and a statistical t value of 23.111 compared to a t-table of 1.96. Because the t-statistic value > t-value (23.111 > 1.96) it can be concluded that perceived ease of use has a positive and significant effect on behavioral intention. So, the hypothesis is accepted. This means that the higher the perceived ease of use, the higher the behavioral intention. Perceived ease of use gives an indication related to a system that the system provides a design to make it easier and not make it difficult for the user. If something is felt by someone to be easier to use then they will be encouraged to have the intention to use it because they have not bothered to run something with high effort. This is supported by the results of distributing questionnaires regarding the variable perceived ease of use which shows that the highest average is found in the statement "Learning to use online health service features on the Halodoc application is easy for me", this identifies that Halodoc application users feel the features Online health service features on the Halodoc application are easy to understand and use. Based on these results, the easier it is for users to understand and use the Halodoc application, the behavioral intention will increase. The results of this study are not in line with previous research show that perceived ease of use has a positive influence on behavioral intention (Prastiawan et al., 2021;Zhang et al., 2020). Meanwhile, the results of similar research show that perceived ease of use does not affect behavioral intention (Cigdem & Ozturk, 2016).
Based on the results of data analysis, it was obtained that the p-value of the social influence variable on behavioral intention was 0.039 which was compared to a significant value of 0.05. Because the p-value < significant (0.039 < 0.05) with a beta value of 0.216 and a statistical t value of 4.923 compared to a t-table of 1.96. Because the t-statistic value > t-value (4.923 > 1.96) it can be concluded that social influence has a positive and significant effect on behavioral intention. So, the hypothesis is accepted. This means that the stronger the social influence, the higher the behavioral intention. According to TAM theory, one of the factors that can encourage users to use a technology is the influence of its social environment. This social influence can be felt directly in the surrounding environment or through social media owned by users. Social influence can encourage someone's desire to use a technology when it becomes a trend in their social environment. This is supported by the results of distributing questionnaires regarding social influence variables which show that the highest average is found in the statement "I use the Halodoc application because the environment or the surrounding community also use online health services on the Halodoc application", this identifies that Halodoc application users use services online health on the Halodoc application because the environment or the surrounding community also use the service. Based on these results, the more people in the same environment use the Halodoc application, the behavioral intention will increase. The results of this study are in line with previous research showing that social influence has a positive effect on behavioral intention (Prastiawan et al., 2021). The same thing was conveyed by similar research showing that social influence has a positive effect on behavioral intention (Zhang et al., 2020). While the results of previous research show that social influence has no effect on behavioral intention (Bimantara, 2021;Nisa & Solekah, 2022;Nurhaliza, 2022;Wintang & Pasharibu, 2021).
Based on the results of data analysis, it was obtained that the p-value of the perceived usefulness variable for behavioral intention was 0.001 compared to a significant value of 0.05. Because the p-value < significant (0.001 < 0.05) with a beta value of 0.214 and a statistical t value of 29.340 compared to a t-table of 1.96. Because the t-statistic value > t-value (29.340 > 1.96) it can be concluded that perceived usefulness has a positive and significant effect on behavioral intention. So, the hypothesis is accepted. This means that the higher the perceived usefulness, the higher the behavioral intention. Perceived usefulness or usefulness in use according to TAM theory is one of the factors that causes the use of a technology. The benefits of using a technology will cause the intention to use the technology to increase, this is because users tend to be attracted to technology that is easy to use so that it speeds up and makes their performance more efficient. This is supported by the results of distributing questionnaires regarding perceived usefulness variables which show that the highest average is found in the statement "Operating the Halodoc application can increase my productivity level in the health sector", this identifies that Halodoc application users feel when using the Halodoc application the productivity of application users Halodoc in the health sector can improve. Based on these results, the more benefits users get when using the Halodoc application, the behavioral intention will increase. The results of this study are in line with previous research illustrate that there is a positive association between perceived usefulness and behavioral intention (Prastiawan et al., 2021;Zhang et al., 2020). Perceived usefulness is a very important determining factor in motivating the intention of an individual to achieve something. In contrast to the results of similar research who show that perceived usefulness has no effect on behavioral intention (Nisa & Solekah, 2022).
Based on the results of data analysis, it was found that the p-value of the E-trust variable on behavioral intention was 0.407, compared to a significant value of 0.05. Because the p-value > significant (0.407 > 0.05) with a beta value of -0.037 and a statistical t value of 1.043 compared to a t-table of 1.96. Because the t-statistic value < t-value (1.043 < 1.96) it can be concluded that E-trust has no effect on behavioral intention. So, the hypothesis is rejected. This means that high E-trust is not able to have a significant influence on behavioral intention. Trust from a consumer is a belief that service or product providers provide reliability in behaving in such a way that they can fulfill the long-term interests of a consumer. However, the results of distributing questionnaires regarding the E-trust variable showed that the lowest average was found in the statement "I believe that the Halodoc application can maintain the confidentiality of my personal data", this identified that there were Halodoc application users who did not believe that the Halodoc application could protect confidentiality of personal data of Halodoc application users. While the highest average is on the item "I believe in using the Halodoc application because it always provides a good experience", which means that Halodoc application users feel confident using the Halodoc application based on the experience of Halodoc application users in using the Halodoc application. Based on these results it is known that the personal data used in the Halodoc application does not reduce user interest in using the Halodoc application due to the good experience that users have of the Halodoc application, so that e-trust cannot significantly influence behavioral intention. The results of this study are in line with previous research showing that trust has no effect on behavioral intention (Beldad & Hegner, 2018). Meanwhile, the results of similar research shows that e-trust has a positive effect on behavioral intention (Baabdullah, 2018;Jatimoyo et al., 2021;Zhang et al., 2020).
Based on the results of data analysis, it was obtained that the p-value of the variable perceived ease of use for perceived usefulness was 0.003 compared to a significant value of 0.05. Because the p-value < significant (0.003 < 0.05) with a beta value of 0.530 and a statistical t value of 18.884 compared to a t-table of 1.96. Because the t-statistic value > t-value (18.884 > 1.96) it can be concluded that perceived ease of use has a positive and significant effect on perceived usefulness. So, the hypothesis is accepted. This means that the higher the perceived ease of use, the higher the perceived usefulness. Ease of using technology can also make technology users have a positive or negative attitude towards technology. The easier a technology is to use; the more users believe that technology will benefit them. This is in line with the TAM theory which explains that users have an attitude in using technology that depends on the ease of use of the technology. This is supported by the results of distributing questionnaires regarding the variable perceived ease of use which shows that the highest average is found in the statement "Learning to use online health service features on the Halodoc application is easy for me", this identifies that Halodoc application users feel the features Online health service features on the Halodoc application are easy to understand and use. Based on these results, the easier it is for users to understand and use the Halodoc application, the user can also easily assess how the benefits provided by the Halodoc application, so that the easier it is for users to understand and use the Halodoc application, the perceived usefulness will increase. The results of this study are in line with previous research shows that perceived ease of use has a positive effect on perceived usefulness (Dhahak & Huseynov, 2020;Huei et al., 2018;Sarkar et al., 2020). Meanwhile, similar research shows that perceived ease of use does not affect perceived usefulness (Weng et al., 2018;Zhang et al., 2020).
Based on the results of data analysis, it was obtained that the p-value of the social influence variable on E-trust was 0.000 which was compared to a significant value of 0.05. Because the p-value < significant (0.000 < 0.05) with a beta value of 0.437 and a statistical t value of 3,088.839 compared to a t-table of 1.96. Because the t-statistic value > t-value (3,088.839 > 1.96) it can be concluded that social influence has a positive and significant effect on E-trust. So, the hypothesis is accepted. This means that the stronger the social influence, the higher the e-trust. The influence of the social environment in the use of technology plays an important role in the use of this technology. Users who feel that the environment around them uses certain technologies will cause a feeling of trust in these technologies to be useful for them. Therefore, social influence can be said to increase trust in the use of technology. This is supported by the results of distributing questionnaires regarding social influence variables which show that the highest average is I Putu Kresna Saniscara Dewanta Putra / Influence Attitude and Behavioral Intention of the Millenial Generation to Adoption of Telemedicine Platforms in Bali in the New Normal Era found in the statement "I use the Halodoc application because the environment or the surrounding community also use online health services on the Halodoc application", this identifies that Halodoc application users use services online health on the Halodoc application because the environment or the surrounding community also use the service. Based on these results, the more people in the same environment use the Halodoc application, the level of trust in the Halodoc application will also increase because the experience gained by one user can affect the trust of other users. The results of this study are not in line with previous research shows that social influence has a positive effect on e-trust (Della Fitriana et al., 2021;Zhang et al., 2020). Meanwhile, the results of similar research actually show that social influence has no effect on e-trust (Yap & Lim, 2017). Based on the results of data analysis, it was found that the p-value of the social influence variable on behavioral intention through E-trust was 0.386 which was compared to a significant value of 0.05. Because the p-value > significant (0.386 > 0.05) with a beta value of -0.016 and a statistical t value of 1.100 compared to a t-table of 1.96. Because the t-statistic value < t-value (1.100 < 1.96) it can be concluded that E-trust is not able to mediate the influence of social influence on behavioral intention. So, the hypothesis is rejected. This means that high E-trust does not have a significant effect on social influence on behavioral intention. The results of distributing questionnaires regarding social influence variables showed the lowest average in the item "I use the Halodoc application because of the habit of the surrounding environment to use health technology", this identified that there were Halodoc application users who did not use the Halodoc application even though their surroundings used health technology. As well as the respondents' answers to the E-trust variable which shows that the lowest average is found in the statement "I believe that the Halodoc application can maintain the confidentiality of my personal data", this identifies that there are Halodoc application users who feel they do not believe that the Halodoc application can maintain confidentiality Halodoc application user personal data. While the highest average is on the item "I believe in using the Halodoc application because it always provides a good experience", which means that Halodoc application users feel confident using the Halodoc application based on the experience of Halodoc application users in using the Halodoc application. Based on these results it is known that social influence influenced by e-trust cannot significantly influence behavioral intention.
Based on the results of data analysis, it was obtained that the p-value of the variable perceived ease of use towards behavioral intention through perceived usefulness was 0.000 which was compared with a significant value of 0.05. Because the p-value < significant (0.000 < 0.05) with a beta value of 0.113 and a statistical t value of 56.538 compared to a t-table of 1.96. Because the t-statistic value > t-value (56.538 > 1.96) it can be concluded that perceived usefulness is able to mediate positively the effect of perceived ease of use on behavioral intention. So, the hypothesis is accepted. This means that with perceived usefulness, the effect of perceived ease of use on behavioral intention will increase. This is supported by the results of distributing questionnaires regarding the variable perceived ease of use which shows that the highest average is found in the statement "Learning to use online health service features on the Halodoc application is easy for me", this identifies that Halodoc application users feel the features Online health service features on the Halodoc application are easy to understand and use. As well as the results of distributing questionnaires regarding perceived usefulness variables, the highest average was obtained in the item "Operating the Halodoc application can increase my productivity level in the health sector" which means that Halodoc application users feel that when using the Halodoc application the productivity of Halodoc application users in the health sector can increase. Based on these results, the easier it is for users to understand and use the Halodoc application which is supported by the benefits obtained when using the Halodoc application, so that the easier it is for users to understand and use the Halodoc application, the behavioral intention will increase.

CONCLUSION
Based on the results of the research analysis and the results of the discussion in the previous chapter, the conclusions of this study are as follows. Perceived ease of use has a positive and significant effect on behavioral intention. This means that the higher the perceived ease of use, the higher the behavioral intention. Social influence has a positive and significant effect on behavioral intention. This means that the stronger the social influence, the higher the behavioral intention. Perceived usefulness has a positive and significant effect on behavioral intention. This means that the higher the perceived usefulness, the higher the behavioral intention. E-trust has no effect on behavioral intention. This means that high E-trust is not able to have a significant influence on behavioral intention. Perceived ease of use has a positive and significant effect on perceived usefulness. This means that the higher the perceived ease of use, the higher the perceived usefulness. Social influence has a positive and significant effect on E-trust. This means that the stronger the social influence, the higher the e-trust. E-trust is not able to mediate the influence of social influence on behavioral intention. This means that high E-trust does not have a significant effect on social influence on behavioral intention. Perceived usefulness is able to positively mediate the effect of perceived ease of use on behavioral intention. This means that with perceived usefulness, the effect of perceived ease of use on behavioral intention will increase.